The overarching goals of this Project are (1) to utilize our epidemiological understanding of cutaneous melanoma in order to create clinical and population risk estimation and risk stratification and (2) to translate these modeling approaches into patient care in an active clinical setting We will explore different risk modeling approaches to better inform primary and secondary prevention strategies. This will lead t stratification for prevention trials, including chemoprevention when agents become available for clinical applications. Extensive epidemiological data from the Nurses' Health Study (121,700 women followed from 1976 to 2000), Health Professionals Follow-up Study (52,000 men followed from 1986 to 2000), and Nurses' Health Study II (116,000 women followed from 1989 to 2001) provide a rich resource on risk factors and the incidence of melanoma in over 230,000 women and 50,000 men. None of these cohorts has funding to evaluate melanoma risk. Given the unique size and scope of these cohorts, we will examine the ability of risk models to classify individuals with regard to risk of melanoma, and we will evaluate if such a classification of risk can be clinically useful in determining primary prevention strategies and in directing the level of screening surveillance. Based on the above, our specific aims are as follows. We will construct statistical models for melanoma risk, incorporating traditional epidemiological risk factors to stratify population subgroups based on risk We will construct separate models for women (combining data from the HS and NHS II cohorts) and women. We will construct separate models for women (combining data from the NHS and NHS II cohorts) and men. We will evaluate the models with respect to both goodness of fit (i.e., predicting incidence in subgroups) and discriminatory accuracy at the individual level, and will evaluate the additional discriminatory accuracy gained by including additional risk factors, including genetic markers. Using results from Aims 1 and 2 above, we will evaluate the potential public health effectiveness, in terms of reduction in disease burden, associated with various high-risk and population-based primary and secondary prevention strategies. Using results from Aims 1 and 2 above, we will test the hypothesis the individual risk awareness will lead to risk-minimizing behaviors. We will perform these studies at the adult general dermatology clinics affiliated with Harvard Medical School Department of Dermatology (i.e. Massachusetts General Hospital, Brigham and Women's Hospital, Beth Israel Deaconess Hospital).